{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Y2EUDACTQKY67M7C2NAIKRFGU5","short_pith_number":"pith:Y2EUDACT","schema_version":"1.0","canonical_sha256":"c68941805382b1efb3e2d3408544a6a740b3dbe100813e78f2191e5b37017ff3","source":{"kind":"arxiv","id":"2501.18154","version":1},"attestation_state":"computed","paper":{"title":"Mixed-Precision Graph Neural Quantization for Low Bit Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dingyi Zeng, Hongyang Zhao, Malu Zhang, Wanlong Liu, Wenyu Chen, Yichen Xiao","submitted_at":"2025-01-30T05:39:01Z","abstract_excerpt":"Post-Training Quantization (PTQ) is pivotal for deploying large language models (LLMs) within resource-limited settings by significantly reducing resource demands. However, existing PTQ strategies underperform at low bit levels < 3 bits due to the significant difference between the quantized and original weights. To enhance the quantization performance at low bit widths, we introduce a Mixed-precision Graph Neural PTQ (MG-PTQ) approach, employing a graph neural network (GNN) module to capture dependencies among weights and adaptively assign quantization bit-widths. Through the information prop"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2501.18154","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-30T05:39:01Z","cross_cats_sorted":[],"title_canon_sha256":"099bce538c82334a6d6608d6d74b51259aaa228b61ff59985b0b8b1b6c5a5c70","abstract_canon_sha256":"dedbb012e419a332330a1c69173f4af77761eb264168d4a31431c867234058b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:07.899944Z","signature_b64":"u2ogNSd80hIE0l7MOFOBST5+cLjaNuNrufpmq/BrZW31Ab2REocoLTTLWQx3lYmkplXl2tS2SCRErH9KlTfvAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c68941805382b1efb3e2d3408544a6a740b3dbe100813e78f2191e5b37017ff3","last_reissued_at":"2026-07-05T10:07:07.899473Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:07.899473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mixed-Precision Graph Neural Quantization for Low Bit Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dingyi Zeng, Hongyang Zhao, Malu Zhang, Wanlong Liu, Wenyu Chen, Yichen Xiao","submitted_at":"2025-01-30T05:39:01Z","abstract_excerpt":"Post-Training Quantization (PTQ) is pivotal for deploying large language models (LLMs) within resource-limited settings by significantly reducing resource demands. However, existing PTQ strategies underperform at low bit levels < 3 bits due to the significant difference between the quantized and original weights. To enhance the quantization performance at low bit widths, we introduce a Mixed-precision Graph Neural PTQ (MG-PTQ) approach, employing a graph neural network (GNN) module to capture dependencies among weights and adaptively assign quantization bit-widths. Through the information prop"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.18154","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2501.18154/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2501.18154","created_at":"2026-07-05T10:07:07.899533+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.18154v1","created_at":"2026-07-05T10:07:07.899533+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.18154","created_at":"2026-07-05T10:07:07.899533+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y2EUDACTQKY6","created_at":"2026-07-05T10:07:07.899533+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y2EUDACTQKY67M7C","created_at":"2026-07-05T10:07:07.899533+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y2EUDACT","created_at":"2026-07-05T10:07:07.899533+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5","json":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5.json","graph_json":"https://pith.science/api/pith-number/Y2EUDACTQKY67M7C2NAIKRFGU5/graph.json","events_json":"https://pith.science/api/pith-number/Y2EUDACTQKY67M7C2NAIKRFGU5/events.json","paper":"https://pith.science/paper/Y2EUDACT"},"agent_actions":{"view_html":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5","download_json":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5.json","view_paper":"https://pith.science/paper/Y2EUDACT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.18154&json=true","fetch_graph":"https://pith.science/api/pith-number/Y2EUDACTQKY67M7C2NAIKRFGU5/graph.json","fetch_events":"https://pith.science/api/pith-number/Y2EUDACTQKY67M7C2NAIKRFGU5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5/action/storage_attestation","attest_author":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5/action/author_attestation","sign_citation":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5/action/citation_signature","submit_replication":"https://pith.science/pith/Y2EUDACTQKY67M7C2NAIKRFGU5/action/replication_record"}},"created_at":"2026-07-05T10:07:07.899533+00:00","updated_at":"2026-07-05T10:07:07.899533+00:00"}